Cisco Product Forecasting Notes
Cisco Product Forecasting
Fiscal Calendar Overview
Cisco fiscal year is divided into four quarters:
Q1: August to October
Q2: November to January
Q3: February to April
Q4: May to July
Key Metrics
Forecast: Predicted sales or bookings for a product in a given period.
Booking Actuals: Realized sales or bookings during the period.
Accuracy: Measure of how close the forecast is to the actual bookings.
Calculated using:
For product level: MAX(0, 1 - ABSOLUTE(BIAS))
For overall accuracy, weighted average based on product cost.
Types of Forecasts
Demand Planners Forecast: Preliminary predictions made by demand planners for future quarters.
Marketing Teams Forecast: Forecast that incorporates input from marketing teams.
Statistical and ML Forecast: Uses statistical models and machine learning techniques to generate predictions.
BIAS and Accuracy
BIAS: Systematic deviation between the forecasted and actual sales.
Positive BIAS (> 0): Overestimation.
Negative BIAS (< 0): Underestimation.
Formula: (Forecast Booking units - Actual Booking units) / Actual Booking Units
Product Life Cycle (PLC)
Sustaining: Products at market maturity with steady demand.
NPI (New Product Introduction): Newly launched products that are being introduced.
Decline: Products experiencing diminishing demand and sales, prone to obsolescence.
Cost Weighted Ranking
Products ranked based on individual cost and their preliminary forecasts.
Product Insights
SWITCH Enterprise Ultra High: High port count enterprise switches.
SWITCH Enterprise High: High port count PoE switches.
SWITCH Enterprise Low: Compact switches with low port count.
SWITCH Data Center: Designed for data center operations; categorized by high and mid performance.
ROUTER Enterprise: Includes models ideal for small and medium businesses with integrated services.
TRANSCEIVER MODULE: Supports high-speed data transmission.
POWER SUPPLY: Provides power for PoE devices within networks.
ACCESS POINT: Advanced wireless access points for high-density environments.
SERVER: Unified Computing Systems optimized for data center environments.
PROCESSOR: CPU modules for data center applications.
MEMORY: Modules designed to enhance server performance.
Forecast Tables (Examples)
Detailed tables showcasing forecasts and actual bookings for various products across fiscal quarters. These are instrumental for analyzing trends, insights, and adjustments necessary for future quarters.
Example Products:
SWITCH Enterprise High: Seen sustained growth and fluctuations predictively.
PROCESSOR: Low forecast and actuals indicating a decline in demand.
Forecasting Accuracy by Team
Different forecasting accuracy metrics for the demand planning team, marketing team, and statistical/ML teams are analyzed for products, showing discrepancies and the areas of improvement needed.